Machine learning can identify newly diagnosed patients with CLL at high risk of infection

Rudi Agius; Christian Brieghel; Michael A. Andersen; Alexander T. Pearson; Bruno Ledergerber; Alessandro Cozzi-Lepri; Yoram Louzoun; Christen L. Andersen; Jacob Bergstedt; Jakob H. von Stemann; Mette Jørgensen; Man Hung Eric Tang; Magnus Fontes; Jasmin Bahlo; Carmen D. Herling; Michael Hallek; Jens Lundgren; Cameron Ross MacPherson; Jan Larsen; Carsten U. Niemann

doi:10.1038/s41467-019-14225-8

Machine learning can identify newly diagnosed patients with CLL at high risk of infection

Rudi Agius, Christian Brieghel, Michael A. Andersen, Alexander T. Pearson, Bruno Ledergerber, Alessandro Cozzi-Lepri, Yoram Louzoun, Christen L. Andersen, Jacob Bergstedt, Jakob H. von Stemann, Mette Jørgensen, Man Hung Eric Tang, Magnus Fontes, Jasmin Bahlo, Carmen D. Herling, Michael Hallek, Jens Lundgren, Cameron Ross MacPherson, Jan Larsen, Carsten U. Niemann

Department of Mathematics

Bar-Ilan University

Research output: Contribution to journal › Article › peer-review

Abstract

Infections have become the major cause of morbidity and mortality among patients with chronic lymphocytic leukemia (CLL) due to immune dysfunction and cytotoxic CLL treatment. Yet, predictive models for infection are missing. In this work, we develop the CLL Treatment-Infection Model (CLL-TIM) that identifies patients at risk of infection or CLL treatment within 2 years of diagnosis as validated on both internal and external cohorts. CLL-TIM is an ensemble algorithm composed of 28 machine learning algorithms based on data from 4,149 patients with CLL. The model is capable of dealing with heterogeneous data, including the high rates of missing data to be expected in the real-world setting, with a precision of 72% and a recall of 75%. To address concerns regarding the use of complex machine learning algorithms in the clinic, for each patient with CLL, CLL-TIM provides explainable predictions through uncertainty estimates and personalized risk factors.

Original language	English
Article number	363
Journal	Nature Communications
Volume	11
Issue number	1
DOIs	https://doi.org/10.1038/s41467-019-14225-8
State	Published - 1 Dec 2020

All Science Journal Classification (ASJC) codes

General Chemistry
General Biochemistry,Genetics and Molecular Biology
General Physics and Astronomy

Access to Document

10.1038/s41467-019-14225-8

Cite this

Agius, R., Brieghel, C., Andersen, M. A., Pearson, A. T., Ledergerber, B., Cozzi-Lepri, A., Louzoun, Y., Andersen, C. L., Bergstedt, J., von Stemann, J. H., Jørgensen, M., Tang, M. H. E., Fontes, M., Bahlo, J., Herling, C. D., Hallek, M., Lundgren, J., MacPherson, C. R., Larsen, J., & Niemann, C. U. (2020). Machine learning can identify newly diagnosed patients with CLL at high risk of infection. Nature Communications, 11(1), Article 363. https://doi.org/10.1038/s41467-019-14225-8

@article{546a0b7602694311b8da91d81ae11bc5,

title = "Machine learning can identify newly diagnosed patients with CLL at high risk of infection",

abstract = "Infections have become the major cause of morbidity and mortality among patients with chronic lymphocytic leukemia (CLL) due to immune dysfunction and cytotoxic CLL treatment. Yet, predictive models for infection are missing. In this work, we develop the CLL Treatment-Infection Model (CLL-TIM) that identifies patients at risk of infection or CLL treatment within 2 years of diagnosis as validated on both internal and external cohorts. CLL-TIM is an ensemble algorithm composed of 28 machine learning algorithms based on data from 4,149 patients with CLL. The model is capable of dealing with heterogeneous data, including the high rates of missing data to be expected in the real-world setting, with a precision of 72% and a recall of 75%. To address concerns regarding the use of complex machine learning algorithms in the clinic, for each patient with CLL, CLL-TIM provides explainable predictions through uncertainty estimates and personalized risk factors.",

author = "Rudi Agius and Christian Brieghel and Andersen, {Michael A.} and Pearson, {Alexander T.} and Bruno Ledergerber and Alessandro Cozzi-Lepri and Yoram Louzoun and Andersen, {Christen L.} and Jacob Bergstedt and {von Stemann}, {Jakob H.} and Mette J{\o}rgensen and Tang, {Man Hung Eric} and Magnus Fontes and Jasmin Bahlo and Herling, {Carmen D.} and Michael Hallek and Jens Lundgren and MacPherson, {Cameron Ross} and Jan Larsen and Niemann, {Carsten U.}",

note = "Publisher Copyright: {\textcopyright} 2020, The Author(s).",

year = "2020",

month = dec,

day = "1",

doi = "10.1038/s41467-019-14225-8",

language = "الإنجليزيّة",

volume = "11",

journal = "Nature Communications",

issn = "2041-1723",

publisher = "Nature Research",

number = "1",

}

Agius, R, Brieghel, C, Andersen, MA, Pearson, AT, Ledergerber, B, Cozzi-Lepri, A, Louzoun, Y, Andersen, CL, Bergstedt, J, von Stemann, JH, Jørgensen, M, Tang, MHE, Fontes, M, Bahlo, J, Herling, CD, Hallek, M, Lundgren, J, MacPherson, CR, Larsen, J & Niemann, CU 2020, 'Machine learning can identify newly diagnosed patients with CLL at high risk of infection', Nature Communications, vol. 11, no. 1, 363. https://doi.org/10.1038/s41467-019-14225-8

TY - JOUR

T1 - Machine learning can identify newly diagnosed patients with CLL at high risk of infection

AU - Agius, Rudi

AU - Brieghel, Christian

AU - Andersen, Michael A.

AU - Pearson, Alexander T.

AU - Ledergerber, Bruno

AU - Cozzi-Lepri, Alessandro

AU - Louzoun, Yoram

AU - Andersen, Christen L.

AU - Bergstedt, Jacob

AU - von Stemann, Jakob H.

AU - Jørgensen, Mette

AU - Tang, Man Hung Eric

AU - Fontes, Magnus

AU - Bahlo, Jasmin

AU - Herling, Carmen D.

AU - Hallek, Michael

AU - Lundgren, Jens

AU - MacPherson, Cameron Ross

AU - Larsen, Jan

AU - Niemann, Carsten U.

PY - 2020/12/1

Y1 - 2020/12/1

N2 - Infections have become the major cause of morbidity and mortality among patients with chronic lymphocytic leukemia (CLL) due to immune dysfunction and cytotoxic CLL treatment. Yet, predictive models for infection are missing. In this work, we develop the CLL Treatment-Infection Model (CLL-TIM) that identifies patients at risk of infection or CLL treatment within 2 years of diagnosis as validated on both internal and external cohorts. CLL-TIM is an ensemble algorithm composed of 28 machine learning algorithms based on data from 4,149 patients with CLL. The model is capable of dealing with heterogeneous data, including the high rates of missing data to be expected in the real-world setting, with a precision of 72% and a recall of 75%. To address concerns regarding the use of complex machine learning algorithms in the clinic, for each patient with CLL, CLL-TIM provides explainable predictions through uncertainty estimates and personalized risk factors.

AB - Infections have become the major cause of morbidity and mortality among patients with chronic lymphocytic leukemia (CLL) due to immune dysfunction and cytotoxic CLL treatment. Yet, predictive models for infection are missing. In this work, we develop the CLL Treatment-Infection Model (CLL-TIM) that identifies patients at risk of infection or CLL treatment within 2 years of diagnosis as validated on both internal and external cohorts. CLL-TIM is an ensemble algorithm composed of 28 machine learning algorithms based on data from 4,149 patients with CLL. The model is capable of dealing with heterogeneous data, including the high rates of missing data to be expected in the real-world setting, with a precision of 72% and a recall of 75%. To address concerns regarding the use of complex machine learning algorithms in the clinic, for each patient with CLL, CLL-TIM provides explainable predictions through uncertainty estimates and personalized risk factors.

UR - http://www.scopus.com/inward/record.url?scp=85078064654&partnerID=8YFLogxK

U2 - 10.1038/s41467-019-14225-8

DO - 10.1038/s41467-019-14225-8

M3 - مقالة

C2 - 31953409

SN - 2041-1723

VL - 11

JO - Nature Communications

JF - Nature Communications

IS - 1

M1 - 363

ER -

Machine learning can identify newly diagnosed patients with CLL at high risk of infection

Abstract

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this